V. Delpizzo et Jl. Borghesi, EXPOSURE MEASUREMENT ERRORS, RISK ESTIMATE AND STATISTICAL POWER IN CASE-CONTROL STUDIES USING DICHOTOMOUS-ANALYSIS OF A CONTINUOUS EXPOSURE VARIABLE, International journal of epidemiology, 24(4), 1995, pp. 851-862
Background. Non-differential errors in exposure measurements have been
shown to lead to differential misclassification of exposure. As a con
sequence, the common tenet that, in absence of bias, imprecise exposur
e assessment can only bias the risk estimates conservatively does not
necessarily hold. We investigate the effects of exposure measurement e
rrors on the risk estimate and on statistical power. Methods. We used
a computer model that simulates a case-control study. We used both hyp
othetical data and data modelled on empirical measurements of environm
ental magnetic fields exposure. Results. Measurement errors are found
to have a lesser impact on risk estimates and statistical power than w
ould have been the case had misclassification been truly non-different
ial. However, for a given cutpoint, a bias away from the null cannot b
e excluded. The predominant direction of the errors is found to have i
mportant consequences on both the study power and the risk estimates.
Conclusion. When sufficient empirical data are available, computer mod
elling may give a more accurate estimate of the effects of measurement
errors than algebraic corrections.